<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Skills on AiSlackin</title><link>https://aislackin.com/tags/skills/</link><description>Recent content in Skills on AiSlackin</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sun, 06 Sep 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://aislackin.com/tags/skills/index.xml" rel="self" type="application/rss+xml"/><item><title>MoACubed: teaching Hermes which profiles actually work</title><link>https://aislackin.com/posts/2026-09-06-moacubed/</link><pubDate>Sun, 06 Sep 2026 00:00:00 +0000</pubDate><guid>https://aislackin.com/posts/2026-09-06-moacubed/</guid><description>&lt;p&gt;I have been running Hermes with multiple profiles for a while. That gives me useful specialization, but it also creates an uncomfortable question:&lt;/p&gt;&#10;&lt;p&gt;&lt;strong&gt;Which profile should handle a given kind of task?&lt;/strong&gt;&lt;/p&gt;&#10;&lt;p&gt;It is easy to answer that question with vibes. One profile feels sharper. Another writes cleaner code. A third catches edge cases the others miss. Then one impressive run becomes a permanent routing rule.&lt;/p&gt;&#10;&lt;p&gt;That is not evaluation. That is anecdote with a config file.&lt;/p&gt;</description></item><item><title>Argus: the intel agent sheds its cron plumbing</title><link>https://aislackin.com/posts/2026-08-27-personal-intel-agent-upgrade/</link><pubDate>Thu, 27 Aug 2026 00:00:00 +0000</pubDate><guid>https://aislackin.com/posts/2026-08-27-personal-intel-agent-upgrade/</guid><description>&lt;p&gt;When I built &lt;code&gt;personal-intel-agent&lt;/code&gt;, it did one thing: watch a topic, fold duplicate coverage into a single sourced card, and ping only when something materially moved. I later added a second engine for typed metric and price monitoring.&lt;/p&gt;&#10;&lt;p&gt;It now has a better name—&lt;strong&gt;Argus&lt;/strong&gt;—and a cleaner boundary with Hermes Agent 0.21.&lt;/p&gt;&#10;&lt;p&gt;The original version had to carry some of its own scheduler plumbing because cron jobs were effectively stateless. Hermes 0.21 changed that. Cron jobs can now use persistent agent memory, carry the previous run&amp;rsquo;s output forward, skip the LLM when a deterministic source has not changed, keep a bounded per-job notepad, select reasoning effort per job, and deliver directly into a bot&amp;rsquo;s canonical chat.&lt;/p&gt;</description></item><item><title>hermes-skills: a catalog for skills that survive machine rebuilds</title><link>https://aislackin.com/posts/2026-08-19-hermes-skills-catalog/</link><pubDate>Wed, 19 Aug 2026 00:00:00 +0000</pubDate><guid>https://aislackin.com/posts/2026-08-19-hermes-skills-catalog/</guid><description>&lt;p&gt;Hermes Agent skills are just folders with a &lt;code&gt;SKILL.md&lt;/code&gt;, which is exactly why they rot: they end up scattered across profiles and machines, unversioned, unrecorded. This repo is the fix I built for myself — a single maintained collection where every skill is registered in a catalog, installable with one command, and documented well enough that future-me (or anyone) can tell what it does without reverse-engineering the prompt.&lt;/p&gt;&#10;&lt;p&gt;The structure is deliberately boring: a &lt;code&gt;_template&lt;/code&gt; scaffold you copy, a &lt;code&gt;skills/README.md&lt;/code&gt; catalog you register into, and an installer script. Stateful skills keep tracker archives and metric snapshots on the host (under &lt;code&gt;~/.intel/&lt;/code&gt; for Argus), never in the repo; durable user preferences now use Hermes&amp;rsquo;s native persistent memory.&lt;/p&gt;</description></item></channel></rss>